▲ 150 points
back
107 comments
"In vertebrates the capacity for subjective experience is supported by integrated structures in the midbrain that create a neural simulation of the state of the mobile animal in space."
Whoah there! I think this might be another case of semantic creep. It's very easy to conflate consciousness with self-image in the brain, and it's regularly mis-sold in this way by neuroscientists.
I once went to a talk at the Royal Society which was supposed to be about consciousness. The bloke just spoke for an hour about things like optical illusions.
By the same token, if I were to create a processing system for a robot which integrated visual, haptic and auditory feedback to create a representation of the robot in space, we're being asked to credit the robot as somehow being conscious.
> we're being asked to credit the robot as somehow being conscious
How do you differentiate a robot from a philosophical zombie, and this one from a human? We can not be sure that other humans apart from myself (yours truly writing these lines); we only make estimated guesses that, because we share the same physical structures, our experience will also be similar.
Therefore, we could as well extend that presumption to a being sharing structures with similar functions. It may be the case that insect structures are too simple for that (maybe, as you point out, it requires complex coordination among subsystems and not just a spacial model), but at least it provides a lower bound.
> Therefore, we could as well extend that presumption to a being sharing structures with similar functions.
This is more or less the current mainstream stance. I think it's fine to do it for practical purposes, for example deciding that animals cannot be ethically subjected to certain treatments by virtue of the complexity of their nervous systems. But this is reasonable doubt + principle of precaution. I'm fine with that.
What we cannot do is make this presumption and pretend that we are still within the realms of serious science or philosophy. The technical name for this presumption is emergentism, that consciousness somehow arises from the interactions of matter. I believe that the reason why this hypothesis is so popular, and even confused with scientific fact, is that it is the less weird. Avoiding weirdness is a common pitfall in the history of science, it leads to temporary dead-ends.
It's fine to say that science does not support the claim that animals have consciousness, as long as you also agree that science does not support the claim that humans have consciousness.
I don't understand what you are saying about emrgentism being non scientific. Emergentism is the reason the reason that the Universe today has so many properties that it didn't have when it was a miasma of isolated hydrogren or whatever around the time of the Big Bang.
> It's fine to say that science does not support the claim that animals have consciousness, as long as you also agree that science does not support the claim that humans have consciousness.
It doesn't. Can you name the scientific theory that predicts the existence of consciousness? We know it exists from direct experience, nothing else.
> I don't understand what you are saying about emrgentism being non scientific. Emergentism is the reason the reason that the Universe today has so many properties that it didn't have when it was a miasma of isolated hydrogren or whatever around the time of the Big Bang.
In this context the philosophical term "emergentism" is commonly used to refer to the theory of mind, not the general concept of emergence.
But no, emergence is not the cause for anything. Emergence is a human mental model, developed to deal with complexity, for example: "life emerges from chemistry". A brain the size of Jupiter might be able to maintain a mental model of all the fundamental building blocks of chemistry interacting to create complex biological organisms. We don't have a brain the size of jupiter, so we need shortcuts. The Universe clearly becomes more complex the more time passes, having started from a state of zero complexity. This is a direct consequence of the laws of physics, but there is no explanation for why the laws of physics are a certain way. They are just brute facts. If you keep asking "why" you will eventually bump into a wall.
Science does not support the claim that humans have consciousness! You know that you are conscious, but you have no way to falsify the hypothesis that everyone OTHER than you is a philosophical zombie. (If you discover such a test, then congratulations, you are now a scientist up there with Aristotle, Newton and Einstein.)
Maybe I'm confusing something, does your comment imply that consciousness is not the result of matter interacting with other matter?
There is a different way to look at it. If you accept that you meet the definition of consciousness, and can derive a definition of what characteristics you possess that would satisfy the definition for another bein to be similarly conscious to you then that definition is at least rational. True, we may not be able to apply that definition to specific humans, robots, or other beings but the definition is a rational measure of the answer.
For me consciousness is an emergent property of language that possesses the concept of a self as opposed the the external world. Other things like sensory perception or a concept of a body are better described as awareness. This distinction allows us to say that a cockroach or robot are aware but not conscious because they can respond to their environment but not self reflect on that data as they have no narrative structures to use.
By this definition, is consciousness is a subset of awareness? If the aforementioned robot (aware but not conscious) were upgraded with symbol processing (and especially the self-symbol), would it then meet the criteria for consciousness?
Not necessarily a subset. I could imagine a self reflective consciousness that doesn’t experience a lot of sensory stimulus.
You’re missing the point. Most people intuitively define consciousness to include not just self image, but the processing related to it—having wants and desires, and the self-conscious reasoning about them. Having self image is only the first, small step.
Even insects have something like wants and desires - let's call them values. They are necessary in order to assure their survival. Values are derived from reward signals, which are selected by evolution. Values guide actions, and actions protect life and reproduction.
Their nervous systems are not self reflective about them, however, and they certainly don’t perform abstract meta-level reasoning about those wants and desires. Impulse-response is not consciousness, even if involves a simple reflective model of self image.
>Whoah there! I think this might be another case of semantic creep. It's very easy to conflate consciousness with self-image in the brain
But it's very hard to say why consciousness shouldn't be conflated with self-image in the brain!
A conscious agent believes, desires, wants, feels, etc.
A "self-image" is more-or-less just proprioception, and it seems very easy to ditch that and still be conscious.
> A conscious agent believes, desires, wants, feels, etc.
Then, when I just observe how an ant behaves, I can definitely see that he "believes, desires, wants, feels etc."
He sees and/or smells the crumb that I've made while eating a sandwich. He "desires" and "wants" it, runs to it. He "believes" it can lift it and "feels" it can do it. He tries, more than once. Then he has to adjust his belief: the crumb is too heavy. So he gives up trying to lift, now he "believes" he should bite a peace of it. And he does that. Then, he has a new "desire" to carry the bitten-off peace back to the "house" in the hole where other ants are. And he "feels" he can do this, and he will eventually do this. If another ant tries to take the crumb he carries, he will "feel" hat he "has to fight" but he can also "decide" it's "not too interesting" and it's "better to search for another crumb." Now tell me how is that behavior different from a person which is unable to speak, and where you would describe it with "beliefs, desires etc."
Isn't the whole thing with consciousness the very existence of a self-image?
Some sort of proprioception sounds like a prerequisite for conscious existence...
What is consciousness without the concept of self, though? "I think, therefore I am" is the basis of how we view human existence, and there is no reason to not think that the concept of self is just another illusion created by our pattern recognizing brain.
> I once went to a talk at the Royal Society which was supposed to be about consciousness. The bloke just spoke for an hour about things like optical illusions.
Perhaps he was trying to convey that apparent subjective experience is an analogous perceptual illusion.
> By the same token, if I were to create a processing system for a robot which integrated visual, haptic and auditory feedback to create a representation of the robot in space, we're being asked to credit the robot as somehow being conscious.
Perhaps it depends on what you mean by "integrated". Consciousness must serve a functional, adaptive purpose, and if your integration does not perform the same function, then it wouldn't be conscious as we understand it.
> Consciousness must serve a functional, adaptive purpose,
Why? Not all of evolution is adaptive. It's also possible that consciousness is something more fundamental that strongly emerges when the right sort of physical activity takes place. On Chalmers view, it's rich information processing that results in consciousness. Evolution would only provide the means for how that information processing came about in animals. But it could also came about because of Boltzman Brain randomly popped into existence or human design.
I'm not really clear what Chalmers means by 'rich information processing' though, and yes I have read some of his stuff and IMHO he doesn't seem to either. He's super smart and well worth reading, but I happen to disagree on this.
>It's also possible that consciousness is something more fundamental that strongly emerges when the right sort of physical activity takes place
Since brains seem to be physical and are conscious, yes they seem to be performing the right sort of activity but that's pretty much tautological. At least the idea that self image and self awareness are key is a concrete theory that we have a chance of testing.
Maybe it's a sort of self-modeling feedback loop. Our brains model our physical bodies, mental state and the mental state of others. At some point, this modeling becomes sophisticated enough to become recursive. The model includes cruder approximations of those models, etc like being between two mirrors facing each other. The symbolic representations expand beyond our ability to perceive them. We become aware of our awareness. Bang! You're conscious.
> Why? Not all of evolution is adaptive.
Because a) humans have been wildly successful, and b) the fact that our brain is inordinately expensive, so if a new class of non-conscious humans that were equally adept as conscious ones ever arose, they would easily outcompete us for resources. Given our numbers, presumably such non-conscious people would already exist. If so, where are they?
> It's also possible that consciousness is something more fundamental that strongly emerges when the right sort of physical activity takes place.
Depends what you mean by "more fundamental". All of our behaviour emerges when the right sort of physical activity takes place. Our entire brain is filled with "rich information processing", so that's an insufficient qualifier.
Why do you think that consciousness is different from a self image in a neural network which is configured to use some effective learning algorithm?
Of course there are other theories but they are just as unproven as this one.
Yes, before we solve consciousness, we first need to get the nomenclature right.
To be fair, as soon as we go near anything really subjective, the nomenclature seems to fall apart across most disciplines. For obvious reasons, I suppose.
For example, the term "Nous":
I would argue that consciousness is the least subjective phenomenon there is.
I think that the difficulty in defining it comes for the fact that good definitions are so fundamentally simple that people miss their depth. My favorite one is this, and please give it a chance before rejecting it. See if it distinguishes consciousness from all other concepts:
Consciousness is that which cannot be doubted.
to be fair, the claim is, "supported by...". this is not the same as equality.
The map is not the territory.
"Frankly, I find the idea of a bug that thinks offensive!"
For anyone not in the know, this is a line from Starship Troopers.
Well done. Forgot about that movie.
see also: "Insects cannot tell us anything about subjective experience or the origin of consciousness"
The title of that note is, of course, possibly not correct.
> Thus, they fail to make a convincing case that insects can tell us anything about subjective experience or consciousness.
This is not the same as "Insects cannot tell us anything about subjective experience or the origin of consciousness"
Right, the title makes a claim, and their conclusion is that the opposite claim has failed to meet its burden of proof...
Thinking that "because something has not met its burden of proof means that the opposite must be true" is a fallacy.
And the original authors' rebuttal to that and other work:
https://www.pnas.org/content/113/27/E3814
It seems like the assertion (very roughly) is something like: a creature is conscious if it can make a subjective decision about something. It's not a terrible definition, but I feel like I'm missing some nuance.
Is this really saying that consciousness is an efficient solution to basic problems of sensory reafference? Does this mean it is the base from which consciousness arrives or is it merely an attribute of consciousness?
Karl Friston's free energy principle says something similar - https://www.youtube.com/watch?v=NIu_dJGyIQI
> How, why, and when consciousness evolved remain hotly debated topics
Conversely, consider whether or not consciousness as a feature of the world actually evolved at all. That is, consciousness could be an inherent part of reality throughout all of reality. See various versions of panpsychism or idealism. The idea is that energy itself is conscious, and what evolves is the shape or form consciousness takes. There is no place in space-time one can point to and say "this is where the lights came on".
To answer the question of "why" is the same as answering the question "why something rather than nothing?" It's not clear whether a semantic or symbolic expression (which is itself a part of consciousness) can satisfactorily answer it (that is, explain the whole).
Similar article thoroughly discussed previously:
> The brain structures that support subjective experience in vertebrates and insects are very different from each other, but in both cases they are basal to each clade. Hence we propose the origins of subjective experience can be traced to the Cambrian.
Insects are thought to have arisen in the Devonian period, did they not?
Maybe look no further to when human babies become conscious, their first moments that they remember can say a lot about what consciousness is. Having memory is a big part of consciousness.
Kids are conscious way before they're making memories that will last until adulthood.
Not saying SSDs have consciousness however
Lacking any intelligible means to communicate, this is tough to say definitively.
>However, consciousness also gives out somewhere. Plants do not have it.
Stopped reading there.
You think they do? Or just that it's an unreasonably certain assertion? Is there a dividing line? If not plants/animals, then where?
If anyone proved beyond doubt that plants are not conscious & don't make decisions, I sure didn't hear about it. I have little doubt there are many flavors of consciousness out there, and while humans and animals and plants have their differences, they are not as large as some assume.
What I believe is this: We don't know nearly enough about consciousness, how it's made, and what it is, to make statements like that. And when people make the mistake of assuming they know otherwise, anything that follows is likely a pile of ballsack.
I suspect the dividing line, if we assume there is a single one, is pretty fuzzy and subjective. If a tree is cut down, is it still a tree? What about when it's cut into planks? At what point does an acorn become a tree; at it's first leaf, first sprout, first branch?
If we had answers to these kind of questions, the world would be very different to what I've observed.
I can't argue with any of that, and of course these things are all fuzzy. But if plants have anything resembling consciousness, it seems to be so different than what we experience, as to be worth distinguishing.